Claim Missing Document
Check
Articles

Found 1 Documents
Search

Analisis Sentimen Ulasan Pengguna Aplikasi E-Wallet di Indonesia Menggunakan IndoBERT dan Pemodelan Topik Berdasarkan Ulasan Google Play Store Astri Cahyani; Iriani Iriani
Industrika : Jurnal Ilmiah Teknik Industri Vol. 10 No. 4 (2026): Industrika: Jurnal Ilmiah Teknik Industri
Publisher : Fakultas Teknik Universitas Tulang Bawang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37090/g4cf9289

Abstract

The development of the digital economy in Indonesia has driven a significant increase in the use of e-wallets such as DANA, GoPay, and ShopeePay. In the competition for digital financial services, analyzing user perceptions and satisfaction is important for assessing service quality. This study aims to identify sentiment trends and main themes in e-wallet user reviews on Google Play Store using a Natural Language Processing (NLP) approach. The methods used include fine-tuning the IndoBERT model for sentiment classification and the BERTopic algorithm for topic modeling. A total of 42,306 reviews were collected, processed through text preprocessing (cleaning, normalization, and stemming), and then divided into 80% training data and 20% test data. Testing results show that the IndoBERT-base-p2 model achieved 90.3% accuracy with the highest F1-score value in the positive class at 0.948, followed by the negative class at 0.836 and the neutral class at 0.166. Sentiment analysis reveals that ShopeePay has the highest proportion of positive reviews (87.82%), followed by DANA (75.91%) and GoPay (69.73%). Topic modeling shows that positive sentiment is dominated by themes of ease and speed of transactions, while user complaints are mostly related to balances not being credited, login issues, and account verification. The results of this study provide empirical insights for e-wallet developers in improving system reliability and user satisfaction. Keywords: Customer Satisfaction, E-Wallet, IndoBERT, Sentiment Analysis